Author: Michel Langeveld
Date: 12:57:15 09/13/00
Go up one level in this thread
On September 13, 2000 at 12:51:05, Frederic Friedel wrote: >All the talk about fussy doubles has galvanised Mathias Feist into action (it >was not his original favourite function). Now we need to test his algorithms. >Does anyone have a (reasonably sized) database with critical games we can test >it on? Please send it to me. We would need to know which games should and should >not be identified as doubles. I have send you guys a testdatabase with 33 doubles in it (about 100 games) A nice horrordatabase "Crapbase 2.0" is placed on Dann Corbits site: ftp://38.168.214.175/pub/a-openings/ ftp://38.168.214.175/pub/b-openings/ ftp://38.168.214.175/pub/c-openings/ ftp://38.168.214.175/pub/d-openings/ ftp://38.168.214.175/pub/e-openings/ Dann has cleaned as best ad he could with freeware tools. All doubles CB8 will find are OR extra welcome detected games OR errors in the detection algoritm of ChessBase. Another nice trick is to dedupe it with Chess Assistent and look what it finds.... ;-) Then dedupe with CB8 and look for differences... Hint: Try to automate this process in some way so Mathias can check his algorithm very thourougly and descent with very less time... Kind regards, Michel Langeveld
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.